literature-report
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external search results and paper metadata, creating an ingestion surface for potential indirect prompt injection attacks.
- Ingestion points: Search results in .asta/documents/literature/find/ and data from external lookup skills.
- Boundary markers: No specific delimiters or instructions are provided to separate paper content from agent instructions.
- Capability inventory: The skill uses Read for file access, TaskOutput for generating reports, and shell execution for metadata processing.
- Sanitization: The instructions do not specify sanitization or validation of content extracted from literature sources.
- [COMMAND_EXECUTION]: The skill instructs the agent to use a CLI tool for paper metadata retrieval.
- Evidence: Execution of asta papers get CorpusId: to fetch details like title, authors, and BibTeX fields.
- Context: This is an expected use of a specialized tool for citation management within the literature review workflow.
Audit Metadata